|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The LIGO Executive Committee
Agenda for Monday August 28, 2000 will be:
(Meeting time: 10:30 am Pacific Time)
Open meeting 10:30 - 11:30
Special Items:
http://www.ligo.caltech.edu/LIGO_web/sidebar/roster.html
It is updated 2 to 3 times per month. Please use it in lieu of any paper rosters you may possess and use it instead of requesting information from the administrative staff. - GHS
No report this week.
WBS 1.2 LIGO Operations--Administration
From: Ed Chargois <chargois_e@ligo.caltech.edu>
>From: Linda Turner - turner@ligo.caltech.edu>
Web pages for the DCC give simple how-to's for document numbering, easy access to the latest on-line documents, and search capabilities for the DCC database.Take a look. . .
From: Esther Cunningham <esther@ligo.caltech.edu>
Press here for ACCOUNTS PAYABLE HISTORY DATA.
From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
No report this week.
From: Florence Kaufman <fkaufman@ligo.caltech.edu>
From: irena@ligo.caltech.edu (Irena Petrac)
Dorothy Lloyd
Progress Period from 8.11 to 8.24
Accomplishments:
A quarterly progress report will be due
shortly for the quarter ending August 31, 2000. I am tentatively
requesting inputs September 15, but we should probably discuss how best
to fit this in.
The following change request has
been submitted:
| CR-0000014 | WBS 1.2.1 | Additional Support Equipment for Commissioning | D. Shoemaker |
Press for the latest Contingency Needs Projection.
From: Kris Duncan <kris@ligo.caltech.edu>
| CR-000005 | Upgrade Prestabilized Lazers | $215,000 |
| CR-000008 | Cameras and Projection System for LLO | $ 26,000 |
| CR-000009 | Cameras and Projection System for LHO | $ 26,000 |
From: Irena Petrac <irena@ligo.caltech.edu>
No report this week.
From: Ed Jasnow <jasnow@ligo.caltech.edu>
The schedule for the Livingston Staging Building has been finalized. The Invitation for Bid will be issued on September 7, the job walk will be held on September 21, and the bid opening will be on October 12. Advertising for this project will be put in the Baton Rouge and New Orleans newspapers next week.
General Items:
--------------
(F. Raab)
Staging activities continue for installation of the WA4K.
Optics:
---------------------
(D. Cook)
Most of this week has been dedicated to the preparations needed for
the 4K
IFO installations. Assembling the optical lever reciever and transmitter
boxes, finishing out the transmission monitor boxes. Betsy is getting
the
COS components built up and ready for bench alignments as well as helping
with COC processing and staging. I have been working with L. Jones
to
assemble two detailed 4k installation specification documents to be
used to
coordinate the 4k IFO installation. One for the end stations and one
for
the corner station. This will help to organize tasks as we will be
installing all the COC, COS, IOO, componets as well as the support
equipment like the optical levers and transmission monitors, the beam
dumps
and baffles during the same vacuum incursion. It is a check list to
track
the progress of each step of the installation including the preliminary
testing of equipment etc. Assisted B.Butler in setting up a test to
measure
the effects of the IR light on the 'new' vs 'old' OSEMs. He has posted
the
results on the e-log (The new OSEMs are ~2000 times less sensitive
to
1064nM IR).
OPTICS/COC/SEI INSTALLATION: Mark Guenther has spent the last week here debugging CAS software, and implementing changes that Hytec suggested. We have successfully run all 3 BSC CAS systems through a battery of tests, and we feel comfortable using the system, although an issue concerning the lower limit to the acceleration rate remains unresolved. (Thanks to Mark and Joe Hanson working both days last weekend!!) We installed high resolution position indicator on each axis of the Parker translation slides, and have performed preliminary characterization measurements. During the CAS testing, we have been using a loaned LDS-1000 laser autocollimator, in conjunction with our own unit which is a component of the IAS alignment setup. We note that each instrument yields a slightly different reading when placed in identical situations and we are investigating why. (Jonathan Kern)
Vacuum: Liquid nitrogen tanks at LLO are now full in preparation for check out of the cryo-pump controls.
Community Relations: Approximately 60 members of the Gulf Coast Section
of the IEEE toured LLO this week.
Covered elsewhere.
Covered elsewhere.
| Installation
& Commissioning:
Livingston |
Other Science/Engineering
Activities:
Issues/Concerns |
See also the Installation web page
The things we found to be important in acquiring the PRM+ARM lock were:
We were also able to image the beam spot on the ETMs using the frame-grabber. Both spots seemed to be well centered on the optics. We expect this method to allow centering to within a centimeter or less.
The readout electronics is currently being assembled, with signal digitization to occur in rack 1x5. Data will be bundled in events triggered by large cosmic ray showers, with each event consisting of pulse height and time information.
We have made arrangements to continue our PEM audit into a second phase, without the help of UT Brownsville students. Doug Lormand will continue to work on this project; he will be assisted on a part-time basis by LSU undergrad Laura Schlecht.
Measured absolute RFAM levels on the beam from the EOMS. After optimization, all are dI/I < 5 X 10^-6 of carrier. The New Focus 1811 RPFDs saturate at ~ 50 uW of input power, so shot noise and any frequency noise riding around the sidebands is below the noise floor of the measurement. We are currently putting in a high power RFPD to get the shot noise up to a level where we can measure it
###SEW: Late update. With the problems in the alignment of the core optics delaying the closure of the vacuum system, it looks like we will have the opportunity to put the new OSEMs on at least two of the optics in the LVEA. We will attempt this unless it results in unacceptable delays.
Keisha Williams finished her term with nice results and paper.
Measurements characterizing the modified current shunt have been completed.
Two prototype boards of the supressed carrier version of the circuit are stuffed and testing should begin by 8/24. Once bench tests at Caltech are complete a board will be sent to LHO for testing. The other board will be sent to Rai at LLO for additional testing.
I also finished updated the trend generation class to allow trending of multiple channels in a single trend frame and automatic writing of the trend frames. I then used this in a new program that monitors the time jitter of the ADC clocks relative to the GPS time reference. I found that the measured timing jitter is 50ns. Several effects (e.g. basline noise) have not been corrected and may contribute to the measured jitter. Rolf says that a <100ns jitter is expected of the GPS time stamp, so the measurement seems to be sufficiently precise.
Minutes of the third meeting of the
40 Meter Interferometer Technical Advisory Committee
Wednesday, August 16, 2000, LSC meeting at LHO.
Ken Strain (chair) kstrain@physics.gla.ac.uk +44 141
330 5884
Minutes by Alan Weinstein, ajw@caltech.edu,
626-395-6682
- Ken Strain gave a brief resume of the newly proposed
LIGO-II control scheme
(I wasn't taking notes, so I can't do it justice)
- Ken went on to describe plans for a quick test of this scheme
at Glasgow 10m. (ditto).
- AJW presented a host of transparencies (see
http://www.ligo.caltech.edu/~ajw/G000194-00.pdf
).
- Re-statement of objectives of the upgrade,
posing the question of whether the control system
risk
is high enough to warrant 40m upgrade effort.
** There were very few comments made!
AJW remains uncomfortable,
and concerned about drumming up
LSC interest/participation.
- ** All of the key elements of the
newly proposed LIGO-II control scheme
can most likely be prototyped at the 40m
in a straightforward and extrapolatable way.
The smaller PRC at the 40m makes the control matrix
a bit less diagonal, but do-able.
- recent activity (dismantling, building rehab,
vacuum control system, output chamber & seismic
stack) described.
- Preliminary optical layout presented, including 12m mode cleaner,
and fixed steering mirrors and MMT lenses.
** AJW requested help/advice in analyzing noise produced
by these fixed elements; no volunteers.
- Need for active seismic isolation (STACIS) addressed.
** No one offered any wisdom on the analysis (MTTL,
minimum frequency for evaluating
rms motion, etc).
** It comes down to whether the installation
can be effectively and safely
managed.
If Larry Jones says he can do
it, we'll do it!
- Question raised as to appropriate size of core optics
(3", 0.25 kg vs 4", 1.56 kg).
(Use 3" SOS optics for MMT1-3, RM, BS, SM).
** Still under debate, advice welcome!
- Question raised as to choice of flat ITMs or symmetric arms.
** opinions favored flat ITMs. OK!
- Very brief mention of designs for LSC and ASC,
in the context of Jim Mason's control scheme.
Designs will be updated for the new proposed LIGO
II control scheme.
_______________________________
OK, now here's AJW's reconstruction
of the main features of the newly proposed LIGO-II control scheme,
as described by Ken Strain and Peter Fritchel:
- phase-modulated sidebands placed before a 16.5 m mode cleaner
at 9 MHz and at 180 MHz
- two Pockels cells in series
- parasitic modulations at 171 and 189 are not a
problem...
- some control signals using balanced RF sidebands
cancel each other when operated in RSE
mode,
but only one sideband is used when detuned,
so these control signals are present.
- small Michelson l- asymmetry to put 180MHz on a bright fringe
out the asymmetric port, while only a small amount
of 9 MHz
goes there.
- beats between 9 and 180 (double-demodulated)
gives clean separation between l+ and ls DOF.
- Other DOF (L+, L-, l-) derived in the usual way.
- GW signal to be obtained using DC locking:
- let some carrier light come out the dark port,
either by:
- take arms slightly out of resonance
in opposite directions
- tiny (additional) Michelson l- asymmetry
to let carrier out
- They like the fist approach better;
not sure why
- beat carrier against signal sidebands to get GW
signal
- get rid of all RF signals with a short output
mode cleaner
(after picking the light off to control
ls, L-)
- Since only low-noise, arm-filtered light is used,
extracting the signal at DC is not too
noisy.
LASTI (Mason, Smith, Kruzel, MacInnis, Shoemaker, Zucker)
---------------------------------------------------------
SEI installation prep (Smith):
The pier anchor holes are marked on two of the three HAMS. Ordered
the
bushing tool to scarify the concrete floor. The drill should
arrive
thursday and some missing pieces needed to be ordered, which should
arrive Thursday or Friday. Started to review the BSC pier installation
procedure and noted that we need a large jig fixture from Livingston
before the holes can even be located. We should also be receiving
the
springs(left and right epoxy and non epoxy springs from Hanford.)
Our bakeout vacuum oven needs to be put back together(missing scroll
pump) to bake out the spring seats.
After sorting out various confusions, finally the reference cavity
simulation could be run with Cella-supplied time-series obtained
from mechanical simulation of this cavity.
It's found that the error signal peaks for its mechanical motion
correspond to frequency modulation with amplitudes of the order
of a few centiHertz or less.
> Noise curve
Key ingredients of noise sources, seismic, thermal and shotnoise, are
in the model now. Biplab is working to iron out one or two oddities
in the E2E_generated noise curves of W2K one FP system.
He will start generating noise curves for the full W2K LIGO
interferometer soon.
> Lock acquisition
Matt is working at LHO to lock the system.
Biplab is running the simulation to find the effect of the misalignment
on the Lock Acquisition.
> Thermal lensing
Hiro will start working to implement the thermal lensing in the current
optics/field implementation. This will be a minimal work to make the
simulation of the LIGO I realistic.
> alfi
An interface to edit a text with multiple lines has been developed.
This enables to edit complex mathematical expressions in function
parser modules.
> Manual
Hiro started to update the manuals, the primitive list and the getting
started.
> A new programmer coming
A new contract programmer is coming at the beginning of September.
His main task is the improvement of the simulation engine. Ed and
Bruce will still focus on GUI.
Software Systems (Blackburn)
The issue with not being able to start multiple threads in the
dataconditionAPI was solved this week. The command used to poll
the state of a thread was fould to be "blocking" meaning that
when you asked if a thread was still running, the result of asking
was not returned until the thread finished. A polling command was
developed for the genericAPI two years ago and available for use
in the dataconditionAPI so it was used in place of the one that
was developed specifically for call-chains. It immediately worked,
allowing use to start multi-threaded call-chains across multiple
cpus on a Sun 450. Having the multi-threaded capability did
reveal that the handling of the results of threads lacked adequate
indexing to associate results with issuing call-chain commands.
This problem is being fixed in the TCL layer. Multi-threads also
revealed a need for an association with thread-ids in the cases
where threaded call chains throw exceptions. This will require a
minor modification to the SWIG layer of the dataconditionAPI.The problem with sending multiple data sets out threaded sockets
was also solved this week. The problem was identified as a compile
time association with a non-thread safe malloc. The LDAS build
scripts have been significantly improved to properly convey all
the needed flags to build thread safe shared objects.The wrapperAPI documentation was modified this week to emphasis
the indexing nature of the interface which was shadowed by the
unfortunate use of the word template. This should make clearer
understanding of the flexibility of the wrapperAPI.Work on the wrapperAPI has continued to improve its communication
performance. It is now 150 times faster than the original version.
I reported ~100x last week.We have started integrating a new dynamically loaded shared
object for parallel searches into the wrapperAPI. The dso was
developed over the summer as part of a SURF project and is an
implementation to search for damped oscillators and chirps using
simple physics models.The internal compiler error that was occurring when LDAS was
being built at UWM has been traced and fixed.This was a very active week for giving personal tutorials on the
LDAS system to members of the LSC visiting LIGO after the meeting
last week. Five individuals asked for and received presentations
which overviewed LDAS and one more is planned for next week. If
nothing else, this past LSC meeting sparked interest in learning
more about the LDAS design.Jeff Edlund (CIT Undergrad):
I completed a prototype LAL implementation of the Fast Chirp Transform for
an arbitrary number of terms. This new implementation also allows more
flexiblity in choosing the evaluation space. For the next week and a half
I'll be working on parallelizing and testing it.
Hardware Systems (Anderson)
Electrical re-wiring of the CIT LDAS production machine room has been
approved.Migration from the old Ultra10 LDAS software development server (spica)
has begun to a new quad-processor E450 (ldas-sw). The new E450 was
temporarily available as dataconsparc in the CIT ldas-dev network, but
is no longer available as such.A 933MHz PIII test beowulf node from HPC was received and is being returned
due to out-of-spec physical dimensions.A multi-vendor GigE switch interoperability test has been organized
in conjunction with hardware from SRL at Caltech.----- Forwarded message from Greg Mendell -----
Report on Activity at LDAS-WA
1) Detailed specifications have been received about various "enterprise
level" Tape Libraries that might be used by LDAS. These have been used
to update drawings of the Mass Storage Room at LHO, showing the floor
plan including the location of tape libraries. Similear drawings have
been requested from LLO.2) An analysis has begun to investigate the requirements needed for LDAS
to write data to tape using automated scripts during upcoming
engineering runs.
Data Analysis (Lazzarini)
The stochastic background paper Finn and I are writing has led to derivation of an
optimal filter for the case in which there is inter-detector background correlated noise.
This is a refinement of the optimal filter found in the current literature and includes the
cross-spectral power density in a natural way.
General Computing
MIT:
Checked the order that has arrived from SUN and it appears that all
of the
equipment has arrived.
The SUN 450 and a number of the Ultra 10's are near completion for
their s/w
installations. All are having Solaris 8 loaded on them which means
a number of
the units now being used will have to be upgraded in the near future.
The video switch and UPS are setup and working but need to be installed
in the
rack when it arrives.
David S. is getting a few more IP addresses for visitor usage.
A number of the old computers are being turned in and replaced with
the new
Ultra 10's.
Livingston:
General Computing budget issues are being worked on.
Working on a number of maint. contracts and new purchases.
Hanford:
Nothing to report.
CIT:
(Sam)
- Back from vacation. Worked on a switch box and developing some web
pages
concerning GC help for all the CIT LIGO users.
- Fixed a few oracle items.
(Barbara)
- Reviewed the Roster database with Elizabeth Wood and made modifications
per her comments. Hopefully every flavor of visitor and resource
can now
be accommodated. Developed a means for adding departmental
information
and printing the one-page paper directory.
(Suresh)
- Has returned from vacation. Assisted on getting 40M network back
on-line.
(Lisa)
- Covering for Larry while he is at MIT.
- Setting up more Ultra 10's.
- Finishing the install of Cadence. There were some license issues
such as a
missing pkg. and incorrect license sent. Hopefully, all will be resolved
before
the old licenses expire.
- Worked with Larry on trouble shooting a server problem.
- Has found a number of machines (along with Stuart) missing their
default
router information. The machines have been taken care of but it is
something to
watch for.
(Larry)
- Worked a number of network issues in Bridge and went over some of
the issues
with Steve and Alan at the 40M.
- Working some maint. contracts with SUN.
- After discussions with Barry and Gary we've decided to mount the
projectors in
the conference rooms and we will purchase a few portable units to help
accommodate the needs for conferences.
- Working on Ultra 10 installations and testing some of the hardware.
From Riccardo DeSalvo:
All SURFERS leaving. Good luck.
The lab will be emptier next week!
Francesco, Soy
Acoustic emission noise levels checked.
Developing triggering scheme of average HF/LF ratio.
Need to estimate actual noise levels.
Akiteru
After the good IP pendulum new oil bearing pump problems, need
to fix
before arrival of new counterweights.
IP flex joints that were at 9 mm (zero frequency for >200Kg load) have
been modified to 8 mm (obtaining 175 mHz frequency at ~120Kg
nominal
load) and will be modified to 7.3 mm (to obtain zero frequency for
~120Kg load). This will happen after counterweight tests.
James (left, report by Akiteru)
Performed test of double TAMA MGASF, will see results next week,
measurement, as expected, limited by accelerometer sensitivity at ?80
dB.
Brett (left, report by Virginio)
PCB board finished plus program compiled.
Lisa (left, report by Francesco)
Need dedicated R?O station, system presently cannibalised. Achieved
1
um stability in LVDTs.
Ruggero, Giovanni, Virginio
Setup of Virgo’s accelerometers on LIGO IP completed. Starting
Inertial
damping on complete IP plus 2 GASF tower. Will make accelerometer
inertialdamping without including the LVDTs for a first step.
Alessandro, Carlo.
Finally found the problem with the monolithic accelerometer
construction.
After breaking many tools in drilling and tapping the mechanics and
wasting ages in EDM the drill bit corpses out of it, they passed to
the
final milling and immediately discovered the problem. The tools
did not
break because of material hardness (as speculated so far). The
forged
material is full of internal cracks (we paid a bundle for material
certified free of defects from the most prestigious German firm) because
of insufficient forging.
As the defects are in the core, far from the flex joints, we will finish
the parts and temporarily use the accelerometers while we will get
replacements from the material provider (and damages, the material
was
certified).
Susha, Giancarlo
After some corrections, freeing some parameter and inserting the
physical shape and the physical boundary conditions the simulation
starts giving physical responses, still some problems.
From: Helena Armandula <ahelena@ligo.caltech.edu>
Silica Bonding
I met in Stanford with Sheila Rowan and J. Hough.
We discussed surface preparation prior to bonding.
Cleanliness of the surface before bonding seems to be closely tied
to bond strength.
Investigated several cleaning methods:
Cleaning with warm Liquinox; cleaning with Micro 90 at different temperatures;
scrubbing the surface with cerium oxide; scrubbing with calcium carbonate.
All these methods clean the glass efficiently, allowing the water
to sheet-off the surface, however, the scrubbing with cerium oxide and
calcium carbonate achieve the results in minutes instead of the long hour
the detergents take.
We made some bonds at Caltech using calcium carbonate to clean the
surfaces; some minor degradation of the surface was observed in two substrates
out of eight, after cleaning.
Scratches do not affect bonds as long as flatness is preserved.
The bonds will be tested after some curing period.
This scrubbing method looks very promising to clean the flats on the
mirrors because it would be very difficult to clean those surfaces any
other way.
From: Phil Willems <willems@ligo.caltech.edu>
Fused silica fibers/ribbons:
----------------------------
We have measured the dependence of the Young's modulus on temperature
for
Suprasil fused silica by analysis of the free oscillations of a fiber
at
various temperatures. The fractional change in the Young's modulus
is
+1.52e-4/K. This data will be used to design the optimum fiber/ribbon
dimensions for the suspensions in light of the nonlinear thermoelastic
noise. (Phil Willems, Virginio Sannibale)
From: Janeen Hazel <romie_j@ligo.caltech.edu>
Working on Advanced LIGO issues.
Working on 40m Upgrade.
From: Sam Richman <srichman@ligo.mit.edu>
Stiff isolation system (Sam Richman, Wensheng Hua)
We have tested a controller in which the vertical signals are used to subtract out a tilt component from horizontal seismometers at low frequencies (~30 to 500 mHz). The vertical signals can come from either position sensors or seismometers. The tilt cancellation was measured and can be done quite accurately. This has allowed us for the first time to close all of the upper stage loops with position sensors and broadband seismometers.
From MIKE SMITH:
LIGO 2
Scattered Light Requirements:
A preliminary scattered light requirement for LIGO 2 was completed,
based on the LIGO 1 interferometer configuration; the requirement will
be revised to incorporate the LIGO 2 gravity wave signal transfer functions.
Conceptual Design:
Preliminary conceptual designs for the ETM telescope, PO/APS telescope,
suspended arm cavity baffle, and all other pick-off optical path elements
and light control elements were completed.
From PETER KING:
3.1 LIGO II PSL
Some modifications to the
draft PSL schedule have been made to
explicitly call out a few tasks, procurement of electronics components
in
particular.
A start has been made on the PSL development plan.
Characterization of the pump diodes for the Stanford slab amplifier
has
begun. The pump diode assembly consists of 24 30-W pump diodes,
each with
their own thermo-electric cooler (TEC). Each TEC controls the
pump diode's
temperature such that its emission wavelength is optimized for maximum
absorption by the Nd:YAG gain medium. Every pump diode is being
temperature
mapped. Each diode temperature is stored so that when time comes
to run
the amplifier, each diode will be operating at its optimum temperature.
This work is being carried out with Todd Rutherford and Shally Saraf.
For additional information about this report, contact sanders@ligo.caltech.edu